readCsv

The readCsv method will read a CSV file into an in-memory table.

Syntax

Parameters

Parameter	Type	Description
path	String	The file to load into a table.
path	Path	The file to load into a table. Note that compressed files are accepted: paths ending in ".tar.zip", ".tar.bz2", ".tar.gz", ".tar.7z", ".tar.zst", ".zip", ".bz2", ".gz", ".7z", ".zst", or ".tar" will automatically be decompressed before they are read.
stream	InputStream	An InputStream providing access to the CSV data.
url	URL	The URL.
specs	io.deephaven.csv.CsvSpecs	Specifications for how to load the CSV file. See CsvSpecs builder options.

Returns

A new in-memory table from a CSV file.

CsvSpecs builder options

To create the specs argument, call CsvTools.builder, set any of the options below, and then call build to get the configured CsvSpecs.

• allowMissingColumns(boolean): Whether the CSV file can have fewer columns than the header specifies.
• concurrent(boolean): Whether to read the input and parse each column on separate threads.
• customDoubleParser(Tokenizer.CustomDoubleParser): The parser to use for values read as double.
• customTimeZoneParser(Tokenizer.CustomTimeZoneParser): A low-level parser for custom time zone strings. CsvTools.builder sets one that understands Deephaven time zone names.
• delimiter(char): The field delimiter character (the character that separates one column from the next).
• escape(Character): The escape character. Use it to read field or line delimiters as literal text, or to read the Java-style escape sequences \b, \t, \n, \r, and \f. It is unset by default.
• fixedColumnWidths(Iterable<Integer>): The column widths of a fixed-width file, when you don't want the library to determine them from the header row.
• from(CsvSpecs): An existing CsvSpecs object whose options are copied into this builder.
• hasFixedWidthColumns(boolean): Whether the input is organized into fixed-width columns rather than separated by a delimiter.
• hasHeaderRow(boolean): Whether the input file has a header row.
• headerLegalizer(Function<String[],String[]>): The function that rewrites column headers into legal names. CsvTools.builder sets Deephaven's column-name legalizer, and setting this option replaces it.
• headers(Iterable<String>): Column headers you supply. If hasHeaderRow is true, these headers replace the headers in the input. If hasHeaderRow is false, they supply the missing headers.
• headerValidator(Predicate<String>): The check that a column header is legal. CsvTools.builder sets Deephaven's column-name validator.
• ignoreEmptyLines(boolean): Whether the library should skip over empty lines in the input.
• ignoreExcessColumns(boolean): Whether the library should allow excess columns in the input.
• ignoreSurroundingSpaces(boolean): Whether to trim leading and trailing white space from non-quoted values.
• nullParser(Parser<?>): The parser to use when all values in the column are null.
• nullValueLiterals(Iterable<String>): The strings that mean a null value in any column that has no null value literals of its own. The default is the empty string.
• numRows(long): The maximum number of rows to process.
• parsers(Iterable<? extends Parser<?>>): The parsers that participate in type inference.
• putHeaderForIndex(int, String): A column header that replaces the header at a 0-based column index.
• putNullValueLiteralsForIndex(int, List<String>): The null value literals for the column at a 0-based column index.
• putNullValueLiteralsForName(String, List<String>): The null value literals for the named column.
• putParserForIndex(int, Parser<?>): A parser forced for the column at a 0-based column index.
• putParserForName(String, Parser<?>): A parser forced for the named column.
• putParsersForIndex(int, List<Parser<?>>): A set of parsers forced for the column at a 0-based column index.
• putParsersForName(String, List<Parser<?>>): A set of parsers forced for the named column.
• quote(char): The quote character. Enclose a field in it to read field or line delimiters inside the field as literal text.
• skipHeaderRows(long): The number of rows to skip before reading the header row from the input.
• skipRows(long): The number of data rows to skip before processing data.
• threadShutdownTimeout(Duration): How long to wait for the reader threads to shut down if processing ends abnormally, such as after an exception. The default is 1 minute.
• trim(boolean): Whether to trim leading and trailing white space from inside quoted values.
• useUtf32CountingConvention(boolean): Whether fixed-width column widths are measured in Unicode code points (true) or in UTF-16 code units, which are Java char values (false).

For the full description of each option, see the CsvSpecs.Builder Javadoc.

Examples

In the following example, write_csv writes the source table to /data/file.csv, and read_csv loads the file into a Deephaven table.

In the following example, read_csv is used to load the file DeNiro CSV into a Deephaven table.

In the following example, read_csv is used to load the file DeNiro TSV into a Deephaven table. The second argument, "\t", indicates that the file contains tab-separated values.

Any character can be used as a delimiter. The pipe character (|) is common. In the following example, the second input parameter is used to read a pipe-delimited file into memory.

Related documentation

• CSV import via query
• CSV export via query
• Docker bind mounts
• Javadoc